The aim of the proposed research is theoretical study of reaction paths in proteins on the atomic level of detail. The reaction path approach suggests an approximate description of many time dependent processes which are difficult to study theoretically by other methods. Examples are conformational changes, ligand diffusion in protein matrices, allosteric transitions and enzymatic reactions. The approach will be employed in calculations of reaction rates as well as in studies of structural properties of reaction intermediates. A new technique will be developed for finding reaction paths in multidimensional systems such as macromolecules. We shall investigate several conformational transitions and enzymatic processes. We shall examine also the effect of mutations and different substrates on the structural properties of intermediates and on the energy profile of the reaction. This part of the study will be performed in collaboration with the experimental group of D. Ringe from MIT, employing their data on the reaction path of aspartate aminotransferase. The experiment focuses on the determination of the structures along the reaction path using mutants and inhibitors at different steps of the reaction. We shall calculate the complete energy profile and study factors determining the energy barriers. The long term goal of the proposed research is to provide automated theoretical technique to find the structural properties and the energetics of reaction paths, and the associated reaction rates. This systematic approach does not suffer from time scale limitations of the usual molecular dynamics techniques. Molecular level studies of protein dynamics on the expanded time scale can advance significantly the understanding of their function. This technique is also expected to be a useful tool in protein design, providing information on the properties of the barriers, where the enzymatic activity is most effective.